tab (version 3.1.2)

tabmedians: Generate Summary Tables of Median Comparisons for Statistical Reports

Description

This function compares the median of a continuous variable across levels of a categorical variable and summarizes the results in a clean table for a statistical report.

Usage

tabmedians(x, y, latex = FALSE, xlevels = NULL, yname = NULL, quantiles = NULL, 
           quantile.vals = FALSE, parenth = "iqr", text.label = NULL, parenth.sep = "-",
           decimals = NULL, p.include = TRUE, p.decimals = c(2, 3), p.cuts = 0.01, 
           p.lowerbound = 0.001, p.leading0 = TRUE, p.avoid1 = FALSE, 
           overall.column = TRUE, n.column = FALSE, n.headings = TRUE, 
           bold.colnames = TRUE, bold.varnames = FALSE, variable.colname = "Variable", 
           print.html = FALSE, html.filename = "table1.html")

Arguments

x

Vector of values for the categorical variable.

y

Vector of values for the continuous variable.

latex

If TRUE, object returned is formatted for printing in LaTeX using xtable [1]; if FALSE, formatted for copy-and-pasting from RStudio into a word processor.

xlevels

Optional character vector to label the levels of x, used in the column headings. If unspecified, the function uses the values that x takes on.

yname

Optional label for the continuous y variable. If unspecified, variable name of y is used.

quantiles

If specified, function compares medians of the y variable across quantiles of the x variable. For example, if x contains continuous BMI values and y contains continuous HDL cholesterol levels, setting quantiles to 3 would result in median HDL being compared across tertiles of BMI.

quantile.vals

If TRUE, labels for x show quantile number and corresponding range of the x variable. For example, Q1 [0.00, 0.25). If FALSE, labels for quantiles just show quantile number (e.g. Q1). Only used if xlevels is not specified.

parenth

Controls what values (if any) are placed in parentheses after the medians in each cell. Possible values are "none", "iqr" for difference between first and third quartiles, "range" for difference between minimum and maximum, "minmax" for minimum and maximum, "q1q3" for first and third quartiles, or "ci.90", "ci.95", or "ci.99" for confidence intervals for the medians (based on binomial probabilities if one or more groups have n less than 10, otherwise based on normal approximation to binomial).

text.label

Optional text to put after the y variable name, identifying what cell values and parentheses indicate in the table. If unspecified, function uses default labels based on parenth, e.g. Median (IQR) if parenth is "iqr". Set to "none" for no text labels.

parenth.sep

Optional character specifying the separator for the two numbers in parentheses when parenth is set to "minmax" or "q1q3". The default is a dash, so values in the table are formatted as Median (Lower-Upper). If you set parenth.sep to ", " the values in the table would instead be formatted as Median (Lower, Upper).

decimals

Number of decimal places for values in table. If unspecified, function uses 0 decimal places if the largest median (in magnitude) is in [1,000, Inf), 1 decimal place if [10, 1,000), 2 decimal places if [0.1, 10), 3 decimal places if [0.01, 0.1), 4 decimal places if [0.001, 0.01), 5 decimal places if [0.0001, 0.001), and 6 decimal places if [0, 0.0001).

p.include

If FALSE, statistical test is not performed and p-value is not returned.

p.decimals

Number of decimal places for p-values. If a vector is provided rather than a single value, number of decimal places will depend on what range the p-value lies in. See p.cuts.

p.cuts

Cut-point(s) to control number of decimal places used for p-values. For example, by default p.cuts is 0.1 and p.decimals is c(2, 3). This means that p-values in the range [0.1, 1] will be printed to two decimal places, while p-values in the range [0, 0.1) will be printed to three decimal places.

p.lowerbound

Controls cut-point at which p-values are no longer printed as their value, but rather <lowerbound. For example, by default p.lowerbound is 0.001. Under this setting, p-values less than 0.001 are printed as <0.001.

p.leading0

If TRUE, p-values are printed with 0 before decimal place; if FALSE, the leading 0 is omitted.

p.avoid1

If TRUE, p-values rounded to 1 are not printed as 1, but as >0.99 (or similarly depending on values for p.decimals and p.cuts).

overall.column

If FALSE, column showing median of y in full sample is suppressed.

n.column

If TRUE, the table will have a column for (unweighted) sample size.

n.headings

If TRUE, the table will indicate the (unweighted) sample size overall and in each group in parentheses after the column headings.

bold.colnames

If TRUE, column headings are printed in bold font. Only applies if latex = TRUE.

bold.varnames

If TRUE, variable name in the first column of the table is printed in bold font. Only applies if latex = TRUE.

variable.colname

Character string with desired heading for first column of table, which shows the y variable name.

print.html

If TRUE, function prints a .html file to the current working directory.

html.filename

Character string indicating the name of the .html file that gets printed if print.html is set to TRUE.

Value

A character matrix with the requested table comparing median y across levels of x. If you click on the matrix name under "Data" in the RStudio Workspace tab, you will see a clean table that you can copy and paste into a statistical report or manuscript. If latex is set to TRUE, the character matrix will be formatted for inserting into an Sweave or Knitr report using the xtable package [1].

Details

If x has two levels, a Mann-Whitney U (also known as Wilcoxon rank-sum) test is used to test whether the distribution of the continuous variable (y) differs in the two groups (x). If x has more than two levels, a Kruskal-Wallis test is used to test whether the distribution of y differs across at least two of the x groups.

Both x and y can have missing values. The function drops observations with missing x or y.

References

1. Dahl DB (2013). xtable: Export tables to LaTeX or HTML. R package version 1.7-1, https://cran.r-project.org/package=xtable.

Acknowledgment: This material is based upon work supported by the National Science Foundation Graduate Research Fellowship under Grant No. DGE-0940903.

See Also

tabfreq, tabmeans, tabmulti, tabglm, tabcox, tabgee, tabfreq.svy, tabmeans.svy, tabmedians.svy, tabmulti.svy, tabglm.svy

Examples

# NOT RUN {
# Load in sample dataset d and drop rows with missing values
data(d)
d <- d[complete.cases(d), ]

# Create labels for group and race
groups <- c("Control", "Treatment")
races <- c("White", "Black", "Mexican American", "Other")

# Compare median BMI in control group vs. treatment group
medtable1 <- tabmedians(x = d$Group, y = d$BMI)

# Repeat, but show first and third quartile rather than IQR in parentheses
medtable2 <- tabmedians(x = d$Group, y = d$BMI, parenth = "q1q3")

# Compare median BMI by race, suppressing overall column and (n = ) part of headings
medtable3 <- tabmedians(x = d$Race, y = d$BMI, overall.column = FALSE, n.headings = FALSE)

# Compare median BMI by quartile of age
medtable4 <- tabmedians(x = d$Age, y = d$BMI, quantiles = 4)

# Create single table comparing median BMI and median age in control vs. treatment group
medtable5 <- rbind(tabmedians(x = d$Group, y = d$BMI), tabmedians(x = d$Group, y = d$Age))
                   
# A (usually) faster way to make the above table is to call the tabmulti function
medtable6 <- tabmulti(dataset = d, xvarname = "Group", yvarnames = c("BMI", "Age"),
                      ymeasures = "median")
                        
# medtable5 and medtable6 are equivalent
all(medtable5 == medtable6)

# Click on medtable1, ... , medtable6 in the Workspace tab of RStudio to see the tables 
# that could be copied and pasted into a report or manuscript. Alternatively, setting the 
# latex input to TRUE produces tables that can be inserted into LaTeX using the xtable 
# package.
# }